저가형 압전 센서를 이용한 주파수 영역 기반 유도전동기 고장 진단 시스템 구현

초록

In this study, a fault diagnosis system for induction motors was developed using a low-cost piezoelectric sensor in the frequency domain. Vibration data were collected from a simulator equipped with induction motors in normal, rotor fault, and bearing fault conditions. The collected data were transformed into the frequency domain using the Fast Fourier Transform (FFT). The resulting frequency-domain vectors were used as inputs to construct a Multilayer Neural Network (MNN) and a 1D-Convolutional Neural Network (1D-CNN), and their classification performances were compared through computer simulations. Furthermore, a real-time fault diagnosis program was implemented using Raspberry Pi and a Graphical User Interface (GUI) based on the trained models. Experimental results confirmed that an effective fault diagnosis system can be achieved even with low-cost hardware platforms.

키워드